APPARATUS AND METHOD FOR PROVIDING PRIMITIVE VISUAL KNOWLEDGE

摘要

Disclosed are a visual knowledge providing device and a visual knowledge providing method. The visual knowledge providing method includes: a step of receiving an input of a video in a digital image sequence form; a step of separating scenes from the inputted video; a step of extracting a representative shot from each separated scene; a step of extracting an object from a frame forming the representative shot; a step of extracting an action verb based on a mutual relation of the extracted objects; a step of selecting the frame in which the mutual relation of the extracted objects which is the basis for the extraction of the action verb is expressed best as a key frame; a step of generating visual knowledge based on the selected key frame; a step of storing the generated visual knowledge in a database; and a step of visualizing the visual knowledge stored in the database to provide the same for an administrator. The purpose of the present invention is to provide the visual knowledge providing device and the visual knowledge providing method capable of forming the visual knowledge by extracting essential information from the video and standardizing the extracted information in the meta data and visualizing the formed visual knowledge.